With this competing renewal, we seek to expand the initial observations of the NYU Women's Health Study showing a strong association between endogenous estrogens (estrone and estradiol) and post-menopausal breast cancer. Of particular interest are subjects who were sampled between 5 and 12 years prior to breast cancer diagnosis. We propose to expand the cohort by 3.5 additional years to identify all subjects developing breast chancer as of the end of 1997. With this additional effort, the cohort will be have been followed for an average of 12 years. We expect to identify a total of 1,548 incident cases of malignant tumors, including approximately 603 cases of invasive breast cancer (257 pre- and 346- menopausal). Of those 124 pre- and 185 post-menopausal cases will have been diagnosed 5 years or more after blood donation. The major aim of the proposal is to examine the association between blood levels of endogenous estrogens (estrone, estrone sulfate, estradiol, bioavailable estradiol), sex-hormone-binding globulin (SHBG) and breast cancer risk among subjects who samples were obtained between 0.5 and 12 years before the date of index diagnosis. We are especially eager to determine whether these associations hold when time to diagnosis increases, i.e., whether the associations are present in the early stages, or even before, disease initiation. We are interested also in determining whether the major androgenic precursors of estrongens (androstenedione and testosterone) are associate with breast cancer risk in the same population. Breast cancer cases and individually matched controls form the cohort will be included in the nested case-control study and their serum samples will be retrieved from storage and analyzed for levels of endogenous hormones utilizing state-of-the-art biochemical methods. Subjects who were pre-or post-menopausal at the time of cohort enrollment (i.e., at the time of the collection of baseline samples) will be considered in separate statistical analyses.